OCRRouter

OCRmyPDF vs OpenOCR

Pick OCRmyPDF to make scanned PDFs searchable and keep them as PDF/A archives. Pick OpenOCR to parse Chinese and English academic PDFs with formulas on your own hardware. The main split is output: OCRmyPDF returns a searchable PDF, while OpenOCR returns text, LaTeX and bounding boxes.

OCRmyPDFocrmypdf/ocrmypdfOpenOCRfudan/openocr
TypeOpen sourceOpen source
Price1 credit / page1 credit / page
Per 1,000 pages$1.00$1.00
Typical speed1.0 s / page0.5 s / page
Best forSearchable PDFs from scansLightweight OCR pipeline on CPU
InputsPDF, imagesPDF, images
Languages100+ languages (via Tesseract)Chinese and English
Max pages per request1,000500
TablesNo✓ Yes
HandwritingNoNo
MathNo✓ Yes
Bounding boxesNo✓ Yes
ConfidenceNoNo
FormsNoNo
Multilingual✓ YesNo
On OCRRouterAvailableAvailable

Prices are what you pay on OCRRouter. Speed is the provider's typical figure, not a benchmark we ran.

Which to choose

Choose OCRmyPDF if

  • You need validated searchable PDF/A output for archiving.
  • You need OCR in more than two languages, since Tesseract language packs cover 100+ languages.
  • You run an office scanner pipeline on CPU-only machines and want work spread across all cores.

Choose OpenOCR if

  • You need formulas recognized and returned as LaTeX, for Chinese and English academic PDFs.
  • You want a small 0.1B-parameter model for document parsing, with 90.57 reported on OmniDocBench v1.5 for OpenDoc-0.1B.
  • You need bounding boxes from text detection and recognition, and you can export to ONNX for deployment.

Key differences

OCRmyPDFOpenOCR
What it doesAdds a Tesseract text layer to scanned PDFsToolkit for text detection, recognition, formula, table and document parsing
Output formatsSearchable PDF (PDF/A by default), plain textPlain text, LaTeX, bounding boxes
LicenceMPL-2.0Apache-2.0
How you run itCommand-line tool, runs locally on CPUSelf-hosted Python library; hardware requirements not published
Languages100+ via Tesseract language packsOpenDoc and UniRec support only Chinese and English
BenchmarkNot published90.57 on OmniDocBench v1.5 (OpenDoc-0.1B), self-reported by the authors with no independent replication found
Known failure--skip-text with default PDF/A output removes form fields; hangs on PDFs whose forms share one Resources dictionaryDocument parsing depends on a separate layout model, PP-DocLayoutV2 from PaddlePaddle

Sources

What each one is

OCRmyPDF

Command-line tool that adds a searchable text layer to scanned PDFs, using Tesseract.

OCRmyPDF details

OpenOCR

Fudan's open toolkit for text detection and recognition, formulas, tables and document parsing.

OpenOCR details

Try both on your own file

Run the file once with each model and compare the output.

PDF, PNG, JPG, WebP, HEIC. Up to 20 MB each, 10 files.No file?

Use one, fall back to the other

{"model": "ocrmypdf/ocrmypdf","fallbacks": ["fudan/openocr"],"file_url": "https://example.com/document.pdf"}

Questions

More comparisons

OCRmyPDF alternativesOpenOCR alternativesBest OCR models in 2026